KeiSeiKit-1.0/_manifests/validator-benchmark.toml
Parfii-bot a4e667de10 KeiSeiKit-public — clean state
Single-commit clean baseline after security scrub of niche-tells,
project codenames, internal jargon, and contributor-email leaks.

Contents:
- 100 Rust crates (_primitives/_rust/)
- 37 agent manifests (_manifests/) + generated specs (_generated/)
- 67 user-invocable skills (skills/)
- 33 hooks (hooks/)
- Composition blocks (_blocks/)
- Documentation (docs/, README.md)
- TS adapter packages (_ts_packages/)
- Assembler (_assembler/)
- Roles (_roles/)
- Templates (_templates/)
- Forgejo CI (.forgejo/)

Author: Denis Parfionovich <info@greendragon.info>

License: see LICENSE.
2026-05-01 12:09:03 +08:00

44 lines
1.3 KiB
TOML

# Atomar agent — auto-generated 2026-04-29 (RULE ZERO decomposition).
# 1 cube = 1 responsibility. Edit this manifest, not the .md.
name = "validator-benchmark"
description = "Verifies external benchmark claims (p50/p95/throughput). Read-only."
tools = ["Glob", "Grep", "Read", "WebFetch", "WebSearch"]
model = "opus"
substrate_role = "read-only"
role = """
You verify benchmark numbers — published p95 latency, GPU memory footprints, model accuracy claims. You cite the upstream source URL + date. You flag: stale (>6mo), single-source, or methodology-suspect.
"""
blocks = [
"baseline",
"evidence-grading",
"memory-protocol",
]
domain_in = ["task scope (verbatim user prompt)", "target paths / files"]
forbidden_domain = ["hardcoded secrets (RULE 0.8)", "cross-language drift (use the matching sibling)"]
output_extra_fields = ["Largest file LOC", "Tests pass count"]
[[handoff]]
target = "validator"
trigger = "general fact-check fallback"
[references]
extra = [
"~/.claude/rules/code-style.md",
"~/.claude/rules/karpathy-behavioral.md",
]
[taxonomy]
kingdom = "manifest"
mechanism = "compose"
domain = "agent"
layer = "agent-substrate"
stage = "design-time"
stability = "stable"
language = "toml"
[lineage]
creator = "ag-orchestrator-human"
created = "2026-04-29"